This proposed research project will lead to creation of an exploratory agent based model (ABM) which will link neighborhood characteristics (i.e. the physical and social environment) with physical activity behaviors in children. There is growing evidence that the physical and social environments exert a strong influence on levels of physical activity in children, an important determinant of childhood obesity. An increasing number of policy interventions targeting physical activity are therefore being proposed. The use of ABM has been recommended by a recent Institute of Medicine committee to help guide policy development targeting childhood obesity. Dr. Joyce Lee, a pediatric endocrinologist and pediatric health services researcher who has expertise in modeling and policy related to childhood obesity, will work with Dr. Ana Diez Roux, an internationally known epidemiologist who has published extensively on neighborhood health effects and the use of agent based models, to create an ABM that will enhance understanding scientific understanding of the ways in which the physical and social environment factors affect physical activity behaviors in children. The specific aims of this study are the following: Aim 1: To perform a comprehensive literature review of: (1) studies which examine neighborhood (physical and social environment) characteristics and their influence on physical activity in childhood and (2) studies that examine the links between physical activity and energy expenditure, and energy expenditure and weight gain among children. Aim 2: To develop an exploratory agent based model which examines the role of neighborhood characteristics (physical and social environment) on physical activity behaviors in children. Aim 2a: To validate and calibrate the model using the data based on nationally representative data and studies from the literature. Aim 2b: To explore the role of spatial inequities in physical education and neighborhood safety on socioeconomic disparities in levels of childhood physical activity. Childhood obesity is a public health crisis, influenced by a complex interplay of factors, including biologic, behavioral, environmental, and social influences. The use of complex systems sciences methodologies (which includes agent based modeling) for enhancing scientific understanding of population health problems is an increasing focus of the National Institutes of Health, with a number of RFAs exclusively focused on this modeling approach. The initial exploratory model from this study will serve as the foundation for a future R01 application to create a comprehensive ABM of childhood obesity which will provide a valuable opportunity to understand how policies at the federal, state and school district levels may impact levels of childhood obesity.